xfs
[Top] [All Lists]

Re: [PATCH 1/8] xfs: merge xfs_qm_export_dquot into xfs_qm_scall_getquot

To: Ben Myers <bpm@xxxxxxx>
Subject: Re: [PATCH 1/8] xfs: merge xfs_qm_export_dquot into xfs_qm_scall_getquota
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Mon, 27 Feb 2012 22:27:40 -0500
Cc: Christoph Hellwig <hch@xxxxxxxxxxxxx>, xfs@xxxxxxxxxxx
In-reply-to: <20120227172310.GH28391@xxxxxxx>
References: <20120220022815.018921977@xxxxxxxxxxxxxxxxxxxxxx> <20120220022903.488870885@xxxxxxxxxxxxxxxxxxxxxx> <20120227015941.GI7762@xxxxxxx> <20120227171727.GA22046@xxxxxxxxxxxxx> <20120227172310.GH28391@xxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Mon, Feb 27, 2012 at 11:23:10AM -0600, Ben Myers wrote:
> On Mon, Feb 27, 2012 at 12:17:27PM -0500, Christoph Hellwig wrote:
> > > I should have replied to this patch, not the older rev.
> > > 
> > > I have a question about the error code, now we're not mapping errors to
> > > EFAULT and we used do... which is correct?
> > 
> > The old one looks finshy, but we even if we changed it it should be
> > in a separate patch with a good description.  I'll resend with this bit
> > fixed.
> 
> Sounds great.  ;)

Actually the patch is fine as-is.  If you look at the old
xfs_qm_scall_getquota we'd return any error we get from xfs_qm_dqget
as-is, and return ENOENT if XFS_IS_DQUOT_UNINITIALIZED is true.  The
place where any other error is converted to EFAULT is never actually
reached.

<Prev in Thread] Current Thread [Next in Thread>